Capstone Experience – 1 course/final project

The Capstone Experience in Data Science (EN.553.806) is a research-oriented project which must be approved by the research supervisor, academic advisor and the Internal Oversight Committee.  The Capstone Experience can be taken in multiple semesters, but the total number of credits required for successful completion is six (6).  Students must complete a Data Science Capstone Experience Proposal form and follow instructions below to submit for approval before enrollment in EN.553.806 will be approved by academic staff.

All students are REQUIRED to present their research findings in poster format at the event held in their final semester. A list of upcoming dates are provided below. Students must also submit a final report to their capstone supervisor. The grade for this course is based, in large part, upon the poster event and your final report. For more information on the poster and the report, please read below.

Semester Last Day of Classes Capstone
Fall 2023 F 12/8 T 11/28
Spring 2024 F 4/26 T 4/16/24
Fall 2024 F 12/6 T 11/26/24
Spring 2025 M 4/28 T 4/15/25
Fall 2025 F 12/5 T 11/25/25
Spring 2026 M 4/27 T 4/14/26

The student must identify and contact a research supervisor who will agree to supervise the capstone experience. The research supervisor must be a JHU faculty member.

The following list includes JHU faculty members who are willing to be contacted by DS students to supervise their capstone project. Click on their name to take you to their webpage so you can learn more about their research interests to see if they align with yours.  This list is not exhaustive and students should feel free to contact other JHU faculty with whom they would be interested to work.

The student will download and complete a Proposal Request for the Capstone Experience in Data Science describing the project goals.

The proposal should be written as follows:

  1. Title of proposed project.
  2. Project description, with enough details for evaluation (e.g., 200 words).
  3. Completion timeline. Be sure to consider adequate time for review and edits by your research supervisor before the end of the semester.
  4. Name and signature of capstone (faculty) supervisor.

TEAM PROJECTS:  Teams (of no more than 2 students) are acceptable to fulfill the capstone experience, with the following requirements:

    • The team submits a single proposal that, in addition to the aforementioned requirements, describes the composition of the group, and indicates how the work is divided among the members of the group. The project must be divided into subtasks and the proposal must indicate which group member will be in charge of each subtask, ensuring that the amount of work expected by each member aligns with the number of credits associated with the capstone (3 credits = 100 hours).
    • Teams should not include more than two students.
    • The poster may be represented individually or as a team.
    • The option to complete a team project is solely at the discretion of the research supervisor.

Student should email the proposal to the project supervisor and supervisor should send signed form back to student.

ANY changes to an approved proposal (including title, team members, title and nature of research) require the student(s) to resubmit a proposal for review.  

  1. Once the proposal has been reviewed and signed by the research supervisor, student will upload the form here. Academic staff will receive notification of the file upload and begin the committee review.
  2. Student should request registration for EN.553.806.01. The approval for the enrollment will not be granted until the student’s proposal has been fully approved.
    • If the deadline has passed for registration, the student must follow the instructions for a late add as indicated on the Registrar’s website.

  1. Student should be prepared to present his/her research findings in the form of a poster presentation which will be scheduled during one of the final dates of the Department Seminar (EN.553.801).
  2. All poster presentations must be IN PERSON.
  3. Presentation Date: Poster presentations take place on Tuesday (at the same time as the Department Seminar), typically 1 week prior to the last day of classes.  For example, in the Fall semester, the presentations are held the last Tuesday of November. In the Spring semester, presentations take place on the last Tuesday of April. Below are upcoming dates for poster presentations:
Semester Last Day of Classes Capstone
Fall 2023 F 12/8 T 11/28
Spring 2024 F 4/26 T 4/16/24
Fall 2024 F 12/6 T 11/26/24
Spring 2025 M 4/28 T 4/15/25
Fall 2025 F 12/5 T 11/25/25
Spring 2026 M 4/27 T 4/14/26
  1. Poster Specs:  The poster should be 24” x 36”.  There is no specific template.  You are encouraged to look at other professional posters of this kind for design ideas. You can also do a quick internet search for Neurips or ICML posters to see examples.

  2. Printing and Cost:  The department recommends FedEx to print posters.  If you use the FedEx on Charles Street, the cost will be covered by the department.  Additional details on contacting FedEx will be providing to you via email.

  1. Each student must write a paper or research report that must be approved by the research supervisor and advisor. The final paper should be 6-12 pages in LaTeX full page format (1 inch margins, times, 12pt) or ms-word equivalent.  Members of team projects must submit individual reports.
  2. Submit your final report to your capstone supervisor no later than 1 week before the last day of classes so that your supervisor has time to review the paper and provide feedback to you. Once any corrections or updates have been made, you will submit your report again to your capstone supervisor for grade. The program coordinator (Lisa Wetzelberger) will contact your supervisor regarding your final grade and will upload those grades to SIS.

At the completion of the project, capstone supervisors will be contacted by the program coordinator (Lisa Wetzelberger) to provide a PASS/FAIL grade which will then be uploaded to SIS.

  • Unix Cluster-Based Training of Large Language Models for User Query Processing
  • Smartwatch Insights: Predictive Health Analytics with Machine Learning
  • Enhancing Sepsis Management: A Machine Learning Approach to Predicting 30-Day Mortality in Sepsis Patients Using the MIMIC-III Database
  • Sports Games Data Extraction with Multiple Object Tracking
  • A model-free controller for supply chain management
  • Analysis of default and prepayment trend of Fannie Mae single family loan
  • Analysis of rare sinonasal tumors
  • Applying causal inference to determine the effect of gender of PI on the safety of clinical trials
  • Bias in college admissions
  • Changes in US methane emissions during the COVID-19 pandemic
  • Data analysis and prediction of successful startups
  • Development of a machine learning model to predict spatially localized gene expression from histology images
  • Online lidar camera calibration using deep learning
  • Self-identification in speech transcription using natural language processing models
  • Stock price prediction using deep learning with feature extension
  • US food desert analysis